build-paths

Top-level router for the build-paths skill set. Asks the user one question (which path?), then dispatches to beginner/, intermediate/, or advanced/. Use when the user wants to scaffold an Oracle-AI-DB project but hasn't picked a difficulty yet.

Quality

81%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

75%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A well-built router skill: it refuses to scaffold, delegates to per-path skills, and gates handoff on explicit prerequisite checks. Its main weaknesses are a triple-repeated API-key explanation and a step-ordering inconsistency (prerequisites check placed after dispatch despite being labeled 'before handoff').

Suggestions

State the OCI_GENAI_API_KEY details (sk-... bearer token, GenAI console, no ~/.oci/config needed, default endpoint) once in Step 3 and reference it from the Stop conditions instead of repeating them three times.

Move the prerequisites check to Step 2 (before the dispatch step) or renumber so the labeled 'before handoff' check no longer follows the handoff step.

Add one line of recovery guidance for failed prerequisite checks (e.g., where to install Docker or point users to OCI docs when a check fails), mirroring the existing stop-condition pattern.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient — a lean router with an exact print block — but the OCI_GENAI_API_KEY explanation (sk-... bearer token, GenAI service console, no SigV1/~/.oci/config) is repeated nearly verbatim in Step 3 items 3-4 and again in the Stop conditions, which could be consolidated into one pass.	3 / 5
Actionability	Fully executable for an instruction-only skill: a copy-paste-ready question block, a dispatch table mapping specific answers ('1 / "beginner" / "easy"') to exact paths, concrete verification commands ('docker --version', 'python --version'), and exact response text for stop conditions.	5 / 5
Workflow Clarity	Steps and validation checkpoints (docker, python, API key) are explicit, but Step 3 is titled 'One-time prerequisites check (before handoff)' yet is numbered after Step 2 'Dispatch', creating a sequencing inconsistency, and there is no recovery guidance if a prerequisite check fails.	4 / 5
Progressive Disclosure	References (build-paths/README.md, build-paths/skills/README.md, PLAN.md) are one level deep, each with a stated purpose, and PLAN.md is conditionally loaded — good discipline. However, no bundle files are present to verify these paths exist, and the ~25-line tier descriptions in the print block overlap content that the referenced README presumably already covers.	4 / 5
	Total	16 / 20 Passed

Description

87%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong description: it answers what and when explicitly, uses third person, and carves out a distinct router niche with no overlap with the tier skills it dispatches to. The only improvement is adding a few more natural synonyms for the Oracle AI DB domain.

Suggestions

Add one or two natural synonyms users would actually say — e.g., 'Oracle 26ai / 23ai' or 'RAG chatbot with Oracle' — to widen trigger coverage without padding.

The two named actions (ask, dispatch) leave specificity just short of the top anchor; a brief mention of the prerequisite check it performs would round out coverage.

Dimension	Reasoning	Score
Specificity	Concrete actions are named — 'Asks the user one question (which path?), then dispatches to beginner/, intermediate/, or advanced/' — with only a minor coverage gap given the skill performs just two actions.	4 / 5
Completeness	Both parts are explicit: what ('Asks the user one question (which path?), then dispatches to beginner/, intermediate/, or advanced/') and when ('Use when the user wants to scaffold an Oracle-AI-DB project but hasn't picked a difficulty yet') with concrete trigger phrasing.	5 / 5
Trigger Term Quality	Good natural keywords ('scaffold an Oracle-AI-DB project', 'hasn't picked a difficulty yet'), but a few natural variations users might say are missing (Oracle 26ai/23ai, RAG, NL2SQL, chatbot).	4 / 5
Distinctiveness Conflict Risk	A clear niche — 'Top-level router for the build-paths skill set' scoped to the not-yet-picked-a-difficulty case — with distinct triggers and minimal risk of firing for the path-specific skills themselves.	5 / 5
	Total	18 / 20 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: oracle-devrel/oracle-ai-developer-hub
Path: build-paths/SKILL.md
Commit: 93ecdd7

Reviewed: 2 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.